Spent today experimenting with Seed Audio 1.0 for podcast production. Instead of editing separate voice, music, and sound effects, I tested generating everything from a single prompt. It saved more time than I expected and made quick iterations much easier. Still exploring new ideas: https://seedaudio1.app/

Spent part of today comparing the same prompt across different AI video models. What surprised me most was how much prompt wording affected the final result. Small changes produced noticeably smoother motion and better scene consistency. I'm keeping Seedance 2.5 in my testing workflow while I continue comparing outputs: https://seedance25.tools/

Spent today improving my AI video workflow with Wan 2.7. I focused on prompt consistency instead of adding new features. After simplifying my prompt templates and tweaking the generation flow, the results became much more predictable with fewer retries. Small changes, but they made a real difference. Still building: https://www.jxp.com/wan/wan-2-7

Working on OmniShow right now. It feels useful for faster AI video concept testing, quick draft generation, and lighter creative iteration before a full production workflow. Still refining the direction, but it already looks practical for storyboarding and early visual exploration. https://omnishow.video/

Working on ERNIE Image right now. It feels useful for faster AI video concept testing, quick draft generation, and lighter creative iteration before a full production workflow. Still refining the direction, but it already looks practical for storyboarding and early visual exploration. https://ernie-image.net/

Working on Wan 2.7 AI Video Generator right now. It feels useful for faster AI video concept testing, quick draft generation, and lighter creative iteration before a full production workflow. Still refining the direction, but it already looks practical for storyboarding and early visual exploration. https://www.jxp.com/wan/wan-2-7